[molpro-user] pi degeneracy with RS2

[Jacky LIEVIN]

Hi,

I'm trying to calculate with RS2 the frequencies of a pi doubly degenerate ground state of a linear molecule. Using "nosym", both components belong to the same symmetry species 1.
The SA-CASSCF calculation (state,2) reproduces correctly the two-fold degeneracy but the RS2 calculation fails with the error message:

Inconsistent CI coefficients in multi and rs2:                  2338
 -2.720941298592908E-003  5.927415631843153E-015
 Check CI vectors in log file!
 
 This error can happen if states are nearly degenerate or for some other reason the Davidson procedure converges slowly.
 One of the following options might help to avoid this problem:
    (i) Use a tighter energy threshold, e.g. GTHRESH,ENERGY=1.d-8
   (ii) Set in the RS2 input OPTION,NSTATI=nstate, where nstate is (at least) the number of states optimized in multi.
  (iii) Increase the pspace thresholds in multi and rs2, e.g. PSPACE,1

None of the above molpro suggestions work.
It seems that RS2 swaps from one degenerate component to the other.
Introducing a slight bent in the geometry solve the problem, but this is not the solution I'm looking for...

Some idea for overcoming the swapping?
Thanks in advance

Jacky
_____________________________________
Prof. Jacky Liévin
Service de Chimie Quantique et Photophysique
Université Libre de Bruxelles, CPi 160/09
50 av F.D. Roosevelt
B-1050 Bruxelles
Belgium
Tel: +32-2-650 4089
Fax: +32-2-650 4232
_____________________________________